|Publication number||US5721918 A|
|Application number||US 08/597,342|
|Publication date||Feb 24, 1998|
|Filing date||Feb 6, 1996|
|Priority date||Feb 6, 1996|
|Also published as||EP0790558A1|
|Publication number||08597342, 597342, US 5721918 A, US 5721918A, US-A-5721918, US5721918 A, US5721918A|
|Inventors||Sven I. Nilsson, Zoran Todorovic|
|Original Assignee||Telefonaktiebolaget Lm Ericsson|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (20), Non-Patent Citations (2), Referenced by (146), Classifications (11), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The present invention relates to a method and system for database recovery, and more particularly, to recovering a primary store database where at least a portion of the contents of the database fluctuate frequently over time and/or where large volumes of data have to be recovered.
From the standpoint of an operating system, a database system carries out transactions to manage the creation, deletion, and modification of "records" stored in the database. One example of a transaction in a mobile telephony application is registering a new cell location of a "roaming" mobile radio telephone, i.e., the radio has moved from geographical cell area 1 to geographical cell area 2. More formally, a database transaction describes a specified sequence of operations that begins with what is sometimes referred to as a "begin" operation and ends with what is sometimes referred to as a "commit" operation or possibly with a "roll back" or "abort" operation. A commit operation signals successful termination of the transaction; a roll back/abort operation signals unsuccessful termination of the transaction. Accordingly, a transaction either succeeds or fails. If the transaction fails, nothing in the database should be changed.
In transaction-oriented database management systems, a recovery manager maintains the "atomic" nature of transactions and reestablishes system operation upon system failures. Failures may have their source in hardware or software. Examples of failures might include failure of the memory medium on which the database is recorded, failure of the computing system in which the database management system is operating resulting for example from a programming error, or failure of one or more transactions to successfully complete processing. Some failures, such as the loss of power, result in loss of knowledge by the system as to its own state, the state of processes under its control, and information with respect to changes/updates being made to the database.
In order to recover from failures, database management systems generally provide a checkpoint and journal/log technique. A checkpoint triggers the copying of the database contents from non-persistent media like random access memory (RAM) onto persistent media such as a magnetic disk at periodic but usually infrequent intervals of time. This periodic copying is sometimes referred to as a "dump." A mechanism is required to ensure that database transactions which occur between checkpoint "dumps" to persistent media are recorded "serially" on a journal or log. The journalled or logged data is sometimes referred to as "delta" information. The logged transactions may be stored in blocks in non-persistent storage buffers, and when the log buffers are full or at the end of relatively short time intervals, the logged transactions are written into persistent media.
While the contents of the database and log buffers in the primary, non-persistent media store may be lost during a system failure, the information stored on persistent media is usually not damaged and is used to restore most of the database information. Those database transactions in progress at the time of the failure which were not logged must be "rolled back" because they were not fully completed. In order to identify which transactions to roll back, the log/journal is searched so that at system restart time, the recovery manager can determine both the transactions that need to be "undone" to effect "roll back" and the transactions that need to be "redone" to effect "commit" in order to serially restore the database to a correct and consistent state.
In practice, the recovery manager may start with two lists: an undo list and a redo list. The undo list initially contains all logged transactions active at the time of the most recent checkpoint, and the redo list is initially empty. The recovery manager serially searches forward through the journal/log starting from the checkpoint record. If the recovery manager finds a "begin" transaction record for a given transaction, it adds that transaction to the undo list. If the manager finds a "commit" record for a given transaction, it moves that transaction from the undo list to the redo list. When the recovery manager reaches the end of the log, the undo list and the redo list identify, respectively, those transactions that must be undone and those which must be redone. Thereafter, the recovery processor goes through the log, serially redoing the transactions on the redo list and undoing the transactions on the undo list.
A significant problem with this recovery procedure is that new transactions cannot be accepted during the time that the database is being recovered. Until the recovery process is complete, many update transactions are lost. Another significant problem for systems with large amounts of data stems from the infrequent occurrence of database image dumps from non-persistent media to persistent media. Thus, the number of recovery log records that must be maintained to recover forward from a most recent image dump/checkpoint grows with time and can become quite large. The larger the log, the longer it takes to recover from a system failure.
A number of patents relate to methods for trying to obtain consistency between the checkpoints and the log including, for example U.S. Pat. Nos. 4,507,751; 4,868,744; 5,278,982; 5,043,866; 5,333,303; 5,386,554; and 5,414,840. Most of these patents do not concern themselves, however, with the size of the log or the considerable time it takes to recover from a fault. U.S. Pat. No. 5,278,982 acknowledges the need for an efficient recovery log archiving protocol but addresses this problem using a pseudo-crash recovery technique that filters and discards most undo log records during the log archiving process. But the filtering of non-essential records takes place after those non-essential records have already been logged.
It is an object of the present invention to reduce the volume of the recovery log archive and the time needed for recovery. Thus, it is an object of the present invention to provide a fast and efficient database recovery of a primary store.
The present invention establishes two different categories of data including semi-permanent data and transient data. Semi-permanent data may include data which change relatively infrequently or not at all. Semi-permanent database transactions may be handled using the conventional checkpoint and logging techniques described, for example, in the above-identified patents. Transient data are typically updated with high frequency. In addition, transient data may be of the type where some inconsistency in that data can be tolerated by the application which uses that data. Database transactions that involve transient data are not logged.
Certain kinds of database applications may be particularly well-suited to classifying data transactions into semi-permanent and transient. For example, consider a mobile telephone database application which contains a database of the location of mobile telephony subscribers. Semi-permanent database records may relate to a relatively static mobile subscriber profile, i.e., the subscriber's identification, his telephone number, services subscribed to, etc. In general, such a subscriber profile changes only infrequently or not at all. However, other more dynamic information relating to that mobile subscriber may be defined by the mobile telephony application as transient data, e.g., the subscriber's current location in the mobile telephony system. When a mobile radio subscriber changes his location, this new location is registered in the database. If such mobile subscriber location changes were logged using conventional logging mechanisms, such logging files would be extremely large. Moreover, given the considerable amount of time required to recover large logging files, it is likely that many subscribers will have already changed their location during that recovery process.
The recovery time in this kind of situation may be quite lengthy, especially where a very large log file is maintained and where the recovery process restores all locations even though only the most recently logged location is of any interest. Since no data transactions are accepted during whatever amount of time the recovery process takes, reducing that recovery time means that fewer transaction updates are lost.
The present invention provides an electronic data storage and processing system where non-persistent memory, such as random access memory (RAM), stores a database divided into first and second memory sections for storing semi-permanent types of data and transient types of data, respectively. A third memory section in RAM is used as intermediate storage, i.e., a logging buffer, for logging database transactions relating only to the semi-permanent data stored in the first memory section of RAM. A first persistent memory, such as magnetic disk, stores a copy of the semi-permanent data stored at an earlier time in the first section of RAM.
At periodic and appropriate checkpoint time intervals, the operating system copies the current data stored in the first section of RAM into the persistent disk memory. In addition, the operating system logs only those database transactions that affect the semi-permanent data stored in the first section of RAM occurring after the most recent checkpoint "dump." In other words, database transactions updating the semi-permanent data that have occurred since the most recent dump are stored in a persistent memory "log." Database transactions affecting the transient type of data stored in the second portion of RAM are not logged. Moreover, the operating system may archive only the semi-permanent data stored in the first section of RAM in the persistent disk memory. However, in an alternative embodiment, even though transient data updates are not logged, the current transient data in the second memory section may also be dumped into the persistent disk at each checkpoint. One way of classifying data as semi-permanent data or transient data is that semi-permanent data usually changes over a time period greater than the checkpoint time interval while the transient data changes more frequently, and oftentimes, during the time period between two checkpoints.
A recovery processor recovers electronic data storage and processing systems from a system failure by copying semi-permanent data from the persistent disk memory over the data currently held in the first section of RAM after the failure. In contrast, the recovery processor leaves the transient data in the second section of RAM in the state that data exists after the system failure. In an alternative embodiment where the transient data are dumped to persistent storage, the transient data in RAM are checked for consistency at reload. If the transient data in RAM are consistent, the transient data in persistent storage are not reloaded. If the transient data in RAM are not consistent, the transient data in persistent storage may be loaded into the second portion of RAM at recovery. That way the transient data are consistent even though possibly outdated.
The present invention also provides a method for managing the data handling system. Database transactions relating to semi-permanent data are stored in a first section of the non-persistent memory. Database transactions relating to transient data are stored in a second section of the non-persistent memory. At predetermined checkpoint time intervals, the data in the first section of non-persistent memory are copied into the persistent backup memory. In an alternative embodiment, the transient data may also be copied into persistent memory at checkpoint intervals and used to recover consistent (but old) transient data should the transient data stored in RAM become inconsistent. Just those database transactions affecting the semi-permanent data stored in the first section of non-persistent memory, but which occur after the most recent checkpoint "dump," are stored in a log portion of persistent memory. Database transactions that affect the transient data are not logged.
Considerable time and data processing resources are saved if only the semi-permanent of data are stored in persistent memory and only database transactions affecting the semi-permanent data are logged. With the amount of data that must be copied back into the non-persistent memory from the persistent memory and then updated using the logged database transactions being reduced because the transient data are not involved in this recovery process, the time required to recover from a system failure is considerably reduced. In addition, the shear volume of data that must be logged and archived is decreased which makes for a more efficient data handling system even in the absence of a system failure.
The foregoing, together with other features and advantages of the present invention, will become more apparent when referring to the following specification, claims, and the accompanying drawings.
For a more complete understanding of the invention, reference is now made to the following detailed description of the embodiments illustrated in the accompanying drawings wherein like reference numerals refer to like elements throughout the drawings:
FIG. 1 illustrates a data handling and storage configuration for storing, updating, logging, and selectively archiving different types of data;
FIG. 2 illustrates a recovery process for the data handling and storage configuration shown in FIG. 1;
FIG. 3 is a function block diagram of a data handling and recovery system in accordance with the present invention;
FIG. 4 illustrates different types of database transactions that may have to be dealt with at system failure;
FIG. 5 illustrates a database system failure and recovery time-line;
FIG. 6 is a flow chart diagram illustrating a database transaction handling procedure in accordance with the present invention; and
FIG. 7 is a flow chart diagram of a recovery procedure for recovering failure of a database system in accordance with the present invention.
In the following description, for purposes of explanation not limitation, specific details are set forth such as particular techniques and configurations to provide a thorough understanding of the present invention. However, it will be apparent to one skilled in the art that the present invention may be practiced in other embodiments that depart from such specific details. In other instances, detailed descriptions of well known methods and devices are omitted so as not to obscure the description of the present invention with unnecessary detail.
FIG. 1 shows a data storage and handling configuration for storing, updating, and selectively archiving different types of data. A "primary store" database 10 is a non-persistent memory such as random access memory (RAM) that stores a database. Data stored in non-persistent memory such as RAM is lost when power to that memory is removed even if only for a very short period of time. Primary store 10 is divided into three memory sections 14, 16, and 18. Of course, memory sections could also be separate memories if desired. The first section of memory 14 stores a first type of data such as semi-permanent data. Semi-permanent data may include data which a particular application changes relatively infrequently or not at all. The second section 16 of the primary database 10 stores a different category of data referred to as transient data. In contrast with semi-permanent data, transient data may change (but not necessarily) with higher frequency. However, semi-permanent data and transient data are used throughout this description as non-limiting examples of two categories of data which are handled differently during database updating and system failure recovery procedures. Of course, other categories of data are also encompassed by this invention. An optional but preferable third section 18 of the primary store database 10 is a log buffer for intermediate storage of database transaction updates that occur after a most recent checkpoint time interval. The log buffer 18 is employed to make the input/output transfer of transaction updates from RAM to disk more efficient.
At each checkpoint time interval, the contents of the semi-permanent data stored in the first memory section 14 are "dumped" or copied into persistent storage media 20 which may be, for example, a magnetic disk or tape. In addition, the information in the log buffer is transferred (preferably frequently) into a persistent memory log 22 which may also be magnetic disk or tape. Significantly, the transient data in the second memory section 16 are in one embodiment of the invention not dumped or copied to any persistent storage media. Instead, transient data may have been determined by an application using the database to change with sufficiently high frequency that it is not effective or efficient to dump that transient data into persistent storage because it is likely that the values of that transient data will change soon after the dump occurs. In other words, to effectively "back up" this transient data so that it could be used to restore the database in the event of a system failure, the backup/dump interval would have to be unrealistically small compared to the checkpoint dump intervals used for the semi-permanent and log buffer data.
Another significant feature shown in FIG. 1 is that although database records in all three sections of the primary store 10 are regularly updated by database transactions, only the update transactions that affect the semi-permanent data in memory section 14 are logged in log buffer 18. Transient data transactions are simply recorded in the second section of the database 16 and are not logged. In this way, the amount of data which is logged in the log buffer 18 is considerably smaller than it would be if all database transactions were logged.
In an alternative embodiment described further below, the transient data storage and RAM portion 16 may be dumped together with the semi-permanent data stored in RAM portion 14 into persistent storage media 20 at the periodic checkpoint intervals. However, in both embodiments, transient data updates are not logged. When the system recovers, the semi-permanent data are reloaded from persistent storage media 20 into semi-permanent data portion 14 of RAM. This feature may be useful for example in situations where the transient data do not change particularly fast.
FIG. 2 illustrates a system failure recovery operation in accordance with the invention shown in FIG. 1. Assuming a system failure, (e.g., a software failure such as a program instruction fault, a hardware failure, or loss of power), the database recovers by preferably resetting or clearing the first memory section 14. Thereafter, the "image" copy of the semi-permanent data stored in the persistent storage media 20 at the most recent checkpoint dump is reloaded into the first section 14. Of course, the first memory section 14 need not be cleared before the reload. In this embodiment, the transient data in memory section 16 are not reset/cleared; nor is any data reloaded into memory section 16. In short, the transient data in memory section 16 are left untouched (and inconsistent in the case of a power failure). The log data from persistent log 22 are used to perform undo and redo operations necessary to recover semi-permanent data transactions in section 14 in accordance with well known procedures outlined in the background and summary section above. Although persistent storage media 20 and persistent log 22 are shown as separate memories for purposes of explanation, both can be implemented using only one persistent memory.
By not having to reload image data into the transient data second memory section 16 and/or by not having to perform the redo and undo operations for transient data based on logged transaction updates, the database recovery time in accordance with the present invention is considerably reduced. As a result, the number of unaccepted transaction updates which occurred during the recovery process are minimized. This minimal recovery time is particularly important for transient data which may be updated quite frequently, and therefore, whose most recent transactional updates are more likely to be lost during a longer database recovery.
In an alternative example embodiment mentioned above, the transient data stored in RAM portion 16 are dumped together with the semi-permanent data stored in RAM portion 14 into persistent storage media 20 at the periodic checkpoint intervals (shown as a dashed line in FIG. 1). However, transient data updates are not logged. When the system recovers, the semi-permanent data are reloaded from persistent storage media 20 into the semi-permanent data section 14 of RAM. The transient data stored in RAM portion 16 are then checked to see if they are consistent. If those transient data in RAM are determined to be inconsistent, the transient data image stored in persistent storage media 20 are reloaded into the transient data portion 16 of RAM (shown as a dashed line in FIG. 2). Thus, the restored transient data would be old but consistent.
In this alternative embodiment, an application programmer has the option to select whether he desires to have the transient data dumped and/or restored. While this embodiment may require some additional time if the transient data in RAM section 16 is determined to be inconsistent upon recovery, there is the advantage of at least having consistent (but old) transient data in the database. However, this embodiment, like the above-described embodiment, saves considerable time at system recovery by not having to perform the redo and undo operations for transient data based on logged transaction updates, i.e., by not having to execute a log for transient data updates.
FIG. 3 is a function block diagram of a database management and recovery system in accordance with the present invention. An image dump archive stores on persistent storage media 20 the last copy of the semi-permanent data in the first section 14 of the primary store 10, and optionally the last copy of the transient data in the second section 16 of the primary store 10. The recovery log 22 is also stored on persistent storage media. Both the image dump archive and recovery log are connected to recovery processor 24, operating system 25, primary store 10, and database manager 26. The database manager 26 is connected to one or more application programs 28 which utilize information in the primary store database 10 in performing their applications. The operating system 25 and database manager 26 control the transaction-oriented data handling and storage system. The recovery processor 24 performs the recovery operations outlined with respect to FIG. 2 and informs the application program via the database manager 26 of the status of the database after recovery.
The status of data in the recovered database can either be consistent but old or inconsistent. A consistent status is typically obtained when a recent backup file has been reloaded from the image dump archive 20 and recovery log 22. Alternatively, the database status may be inconsistent if the backup files are relatively old or if the system completely lost power. The consistency of the data may be determined using for example checksum error detection techniques where checksums are calculated at each dump and at reload for structure information of the database, e.g., record addresses, numbers of instances, allocated memory, etc. If the checksums match after a reload, the transient data are consistent. However, if the checksums do not match, the recovery processor 24 may inform the application that the database is inconsistent, and the application must restore the transient part of the database using information other than that currently stored in the database.
FIG. 4 illustrates in time-line format various transaction categories which may have to be dealt with in the event of a system failure. Database transactions relating to semi-permanent data entries in the first section of memory are shown as semi-permanent transactions ST1-ST5. At time Tc, a checkpoint memory dump occurs. At Tf, a system failure occurs. Semi-permanent transaction ST1 is completed before the checkpoint Tc, and therefore, is readily recovered when the archive copy of the database is reloaded from persistent storage into the non-persistent database. Semi-permanent transaction ST2 started prior to checkpoint time Tc, completed after the checkpoint time Tc, but before the system failure time Tf. Semi-permanent transaction ST3 also started prior to the checkpoint time but did not complete before the system failure. Semi-permanent data transaction ST4 started after the checkpoint but was completed before system failure. Finally, semi-permanent data transaction ST5 also started after the checkpoint time, but was not completed by the system failure.
From the time-line drawing in FIG. 4, one may understand that uncompleted transactions ST3 and ST5 must be "undone" by the database manager 26. Transactions ST2 and ST4 must be "redone" because even though these transactions were completed before the system failure, they were not output on the most recent dump to persistent storage. Once the recovery processor 24 identifies transactions ST2-ST5, it determines which transactions must be undone and which must be redone using the undo-list and redo-list technique described in the background section above.
FIG. 5 illustrates an example time line of database transactions during a system failure and recovery. At time Tc, a checkpoint copy is made of the semi-permanent data stored in the first memory section 14 of the primary store database 10. Various database transaction updates occur between the checkpoint time Tc and the system failure time Tf. Only the transactions ST1, ST3, and ST5 relating to semi-permanent data are logged. Transient data transactions TT2 and TT4 are not logged.
After the system failure at failure time Tf, the recovery process begins and the system starts to reload. At recovery time Tr, the reload is completed, and the data handling and storage system begins operating again. At start time Ts with the system restarted, the log is executed, and at operation time To, normal operation of the data handling and storage system is restored.
Because the database transaction T6 occurs between failure Tf and start Ts, it is not recorded and therefore is effectively lost. As described above, the time when the system starts to accept transient data transactions is shifted from operation time To to the earlier start time Ts. However, if transient data are updated after the loading the of database copy from the persistent storage archive into the first section of the database, those transient data updates, represented in FIG. 5 by transaction T7, are in fact registered in the transient data, second portion of the database. This earlier registration of new database transactions is significant. If conventional logging type recovery techniques were simply used, database transactions such as transaction T7 which occur between time Ts and To would be lost.
The present invention, by reducing the time period of database recovery relating to copying archived data from persistent memory to non-persistent database, reduces the number of lost transactions that occur during that recovery time period is reduced. In addition, database transactions which relate specifically to rapidly changing transient data may be registered before the "normal" operation of the semi-permanent data portion of the primary store database is restored, i.e., after the undoing, redoing, and log restoration procedures required for semi-permanent data. The size of log files in persistent memory (and in temporary log buffers in primary store 10) may be reduced because transient transactions are not logged.
The present invention is also advantageous in applications like intelligent networks which monitor huge amounts of data. Because this data is typically backed up at a managing site, the intelligent network may order a separate recovery orchestrated by that managing site if it detects that the data (semi-permanent and/or transient) is corrupted. Because of that external backup, it is not necessary for the database to store such large quantities of data on the periodic checkpoint dumps, irrespective of whether the data are semi-permanent or transient.
FIG. 6 illustrates in flow chart format a routine for updating a database in accordance with the present invention. A decision is made at block 52 whether a database transaction update has just been received. If so, a second decision is made in block 54 to determine the type of the just received database transaction. If the type of transaction relates to semi-permanent data, that database transaction is stored in the log buffer section 18 of the primary store 10 (block 56). That transaction is also stored in the first database memory section 14 (block 58). A decision is made in block 60 whether the log buffer is full or if it is time to dump the log to persistent media. If not, control returns to check for receipt of another database transaction. Otherwise, the log information is stored on persistent media (block 64).
If the database transaction relates to transient data, that database transaction is stored in the second memory section 16 of primary store 10 (block 62). A decision is made after execution of blocks 62 or 64 whether it is time for a checkpoint dump (block 66), and if not, control returns to wait for receipt of the next database transaction. Otherwise, the semi-permanent data stored in first memory section 14 (and possibly the transient data in the second memory section 16 in an alterative embodiment) are dumped into persistent media (block 68).
FIG. 7 illustrates a recovery routine (block 80) which may be used to recover the database handling and storage system from system fault or failure in accordance with the present invention. A decision is made in block 82 whether a system failure has occurred. If a failure has occurred, the recovery processor optionally clears/resets the first memory section 14 of the primary store 10 (block 84). However, the second memory section 16 of the primary store 10 for storing transient type data is left untouched. The image copy of the semi-permanent data from persistent storage from the most recent checkpoint dump is copied into the first memory section 14 (block 86). The recovery processor further recovers database transactions relating to semi-permanent data which have occurred since the most recent checkpoint using the log information from persistent log 22 (block 88). The recovery processor checks the transient data in the second section 16 of the primary store 10 (block 90). A decision is made in block 92 whether that data is consistent. If it is, the application continues, and further updates to that data are received (block 94). Otherwise, the data are inconsistent, and must be recovered from sources other than the database system (block 96) or in the alterative embodiment, an image copy of transient data from persistent storage from the most recent checkpoint dump is copied into the second memory section 16).
One particular application of the present invention is to mobile telephony systems that manage a large database of mobile subscribers. The mobile telephony application designates certain database records according to type, e.g., as semi-permanent or transient. In general, a mobile subscriber profile, including the subscriber's identification, his telephone number, additional telephony services to which that particular subscriber subscribes, etc., changes infrequently or not at all. In contrast, a roaming mobile telephone (i.e., a subscriber travelling in an automobile between various cellular telephone "cells") may register a new cell location very frequently. Data relating to a subscriber's current location may be declared by the mobile telephony application as transient data. Thus, when a mobile subscriber changes his location, his new location is simply registered in the transient section of the database but not logged. If such mobile subscriber location changes were logged using conventional logging techniques, the logging files would be huge. Moreover, the amount of time required to recover such large logging files after a system failure would be quite extended with potentially many database transaction updates occurring during that extended recovery process being lost. Those mobiles whose transaction updates are not recorded during extended recovery are then "lost" from the perspective of the subscriber database.
The mobile telephony application in accordance with the present invention recognizes that this mobile location type data is suitable for designation as transient type data because it changes very frequently. Alternatively and/or additionally, data which can be tolerated by the application as inconsistent data may also be appropriately designated as transient. Since the transient data are not dumped or logged, they are assumed to be consistent--a good assumption in certain types of system failures. Otherwise or in addition, the application must handle the inconsistency of that transient data.
In the case of changing mobile telephone locations, mobile subscribers frequently and automatically send in registration messages to the mobile telephony application which may be used by the application to restore current database values or correct data values for inconsistent transient data. Alternatively, the mobile telephony application could broadcast a localization order to all mobile subscribers requiring them to report in their current location which would then be used to restore the database. Another alternative would be to use application files external to the database. Such application files (which may be quite large) can be stored in memory independent of the database system. In this situation, such large amounts of data (which may or may not change rapidly), may be defined as transient because the application has an easy mechanism by which to restore the database in the event a system failure renders that transient data inconsistent, such as the intelligent network example described above. This is another example of how the amount of data which has to be archived and logged by the database system may be reduced using the present invention.
While the invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not to be limited to the disclosed embodiment, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4498145 *||Jun 30, 1982||Feb 5, 1985||International Business Machines Corporation||Method for assuring atomicity of multi-row update operations in a database system|
|US4507751 *||Jun 21, 1982||Mar 26, 1985||International Business Machines Corporation||Method and apparatus for logging journal data using a log write ahead data set|
|US4648031 *||Jun 21, 1982||Mar 3, 1987||International Business Machines Corporation||Method and apparatus for restarting a computing system|
|US4868744 *||Mar 3, 1986||Sep 19, 1989||International Business Machines Corporation||Method for restarting a long-running, fault-tolerant operation in a transaction-oriented data base system without burdening the system log|
|US5043866 *||Apr 8, 1988||Aug 27, 1991||International Business Machines Corporation||Soft checkpointing system using log sequence numbers derived from stored data pages and log records for database recovery|
|US5274802 *||Feb 22, 1991||Dec 28, 1993||Gte Mobilnet Incorporated||Method for restoring lost databases by comparing existing database and generic database, and generating cellular switch commands to update the generic database|
|US5278982 *||Dec 23, 1991||Jan 11, 1994||International Business Machines Corporation||Log archive filtering method for transaction-consistent forward recovery from catastrophic media failures|
|US5287496 *||Feb 25, 1991||Feb 15, 1994||International Business Machines Corporation||Dynamic, finite versioning for concurrent transaction and query processing|
|US5297189 *||May 29, 1991||Mar 22, 1994||Alcatel N.V.||Wireless telephone service subscription data user access method|
|US5303286 *||Mar 29, 1991||Apr 12, 1994||Space Systems/Loral, Inc.||Wireless telephone/satellite roaming system|
|US5333303 *||Mar 28, 1991||Jul 26, 1994||International Business Machines Corporation||Method for providing data availability in a transaction-oriented system during restart after a failure|
|US5375251 *||Jun 11, 1992||Dec 20, 1994||Standard Elektrik Lorenz Aktiengesellschaft||Method for generating a temporary mobile subscriber identity signal (TMSI) from a permanent mobile station identity signal (TMSI)|
|US5386554 *||May 16, 1991||Jan 31, 1995||Kabushiki Kaisha Toshiba||Method and apparatus for reducing data locking time by removing a lock when journal data is written into a main memory journal queue|
|US5414840 *||Jun 25, 1992||May 9, 1995||Digital Equipment Corporation||Method and system for decreasing recovery time for failed atomic transactions by keeping copies of altered control structures in main memory|
|US5457796 *||Jun 10, 1994||Oct 10, 1995||At&T Ipm Corp.||File system with components shared between separately addressable versions|
|US5485608 *||Apr 14, 1994||Jan 16, 1996||Oracle Corporation||Methods and apparatus for updating information in a computer system using logs and state identifiers|
|US5561795 *||May 13, 1994||Oct 1, 1996||Unisys Corporation||Method and apparatus for audit trail logging and data base recovery|
|US5577246 *||Feb 22, 1995||Nov 19, 1996||Lucent Technologies Inc.||Database memory compaction and reclamation method|
|US5581750 *||Mar 15, 1993||Dec 3, 1996||International Business Machines Corporation||System and method for improving data recovery performance|
|US5594942 *||Dec 22, 1995||Jan 14, 1997||Telefonaktiebolaget Lm Ericsson||Restoration of a home location register in a mobile radio system|
|1||*||Proceedings of the International Conference on Data Engineering, Kobe, Japan, Apr. 8 12, 1991, Institute of Electrical and Electronics Engineers, pp. 436 443, XP000325905, Vijay et al., Performance Measurement of Some Main Memory Database Recovery Algorithms .|
|2||Proceedings of the International Conference on Data Engineering, Kobe, Japan, Apr. 8-12, 1991, Institute of Electrical and Electronics Engineers, pp. 436-443, XP000325905, Vijay et al., "Performance Measurement of Some Main Memory Database Recovery Algorithms".|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US5850508 *||Nov 20, 1997||Dec 15, 1998||Electronics And Telecommunications Research Institute||Method of prevention of dangling transaction occurrence using a transaction table initialization technique at an analysis step|
|US6067454 *||Apr 14, 1998||May 23, 2000||Telefonaktiebolaget Lm Ericsson (Publ)||Mobile switching center restart recovery procedure|
|US6119128 *||Mar 30, 1998||Sep 12, 2000||International Business Machines Corporation||Recovering different types of objects with one pass of the log|
|US6122630 *||Jun 8, 1999||Sep 19, 2000||Iti, Inc.||Bidirectional database replication scheme for controlling ping-ponging|
|US6173420||Oct 31, 1997||Jan 9, 2001||Oracle Corporation||Method and apparatus for fail safe configuration|
|US6181937||Jul 1, 1998||Jan 30, 2001||Erkki Joensuu||Method for avoiding unnecessary signalling in a cellular communications system|
|US6195709 *||Mar 2, 1998||Feb 27, 2001||International Business Machines Corporation||Method of providing persistency for transient objects in object oriented technology|
|US6226651 *||Mar 27, 1998||May 1, 2001||International Business Machines Corporation||Database disaster remote site recovery|
|US6263338 *||Jul 15, 1998||Jul 17, 2001||Telefonaktiebolaget Lm Ericsson (Publ)||Method relating to databases|
|US6321234 *||Nov 8, 1996||Nov 20, 2001||Sybase, Inc.||Database server system with improved methods for logging transactions|
|US6490610 *||May 30, 1997||Dec 3, 2002||Oracle Corporation||Automatic failover for clients accessing a resource through a server|
|US6526417 *||Jan 25, 2000||Feb 25, 2003||International Business Machines Corporation||System and method for change accumulation unmerged update reduction|
|US6584477 *||Oct 14, 1999||Jun 24, 2003||Hewlett Packard Development Company, L.P.||High speed system and method for replicating a large database at a remote location|
|US6594676||Apr 10, 2000||Jul 15, 2003||International Business Machines Corporation||System and method for recovery of multiple shared database data sets using multiple change accumulation data sets as inputs|
|US6618851 *||Aug 31, 1999||Sep 9, 2003||Autodesk, Inc.||Method and apparatus for state-reversion|
|US6651073 *||May 23, 2000||Nov 18, 2003||International Business Machines Corporation||Method and apparatus for insuring database data integrity without data recovery logging|
|US6662196||Mar 16, 2001||Dec 9, 2003||Iti, Inc.||Collision avoidance in bidirectional database replication|
|US6687851||Apr 13, 2000||Feb 3, 2004||Stratus Technologies Bermuda Ltd.||Method and system for upgrading fault-tolerant systems|
|US6715102 *||Feb 15, 2000||Mar 30, 2004||Romwin Limited Company||Operation input processing apparatus and method|
|US6728735 *||Mar 12, 2001||Apr 27, 2004||Network Appliance, Inc.||Restartable dump that produces a consistent filesystem on tapes|
|US6728747||May 10, 2002||Apr 27, 2004||Oracle International Corporation||Method and system for implementing failover for database cursors|
|US6732123 *||Feb 23, 1998||May 4, 2004||International Business Machines Corporation||Database recovery to any point in time in an online environment utilizing disaster recovery technology|
|US6742138 *||Jun 12, 2001||May 25, 2004||Emc Corporation||Data recovery method and apparatus|
|US6751636||Dec 5, 2000||Jun 15, 2004||Silicon Graphics, Inc.||System and method for maintaining and recovering data consistency across multiple instances of a database|
|US6766413||Mar 1, 2001||Jul 20, 2004||Stratus Technologies Bermuda Ltd.||Systems and methods for caching with file-level granularity|
|US6785695||Oct 19, 2000||Aug 31, 2004||International Business Machines Corporation||System and method for operational assistance during system restoration|
|US6802022||Sep 18, 2000||Oct 5, 2004||Stratus Technologies Bermuda Ltd.||Maintenance of consistent, redundant mass storage images|
|US6820213||Apr 13, 2000||Nov 16, 2004||Stratus Technologies Bermuda, Ltd.||Fault-tolerant computer system with voter delay buffer|
|US6868506||Mar 12, 2004||Mar 15, 2005||Emc Corporation||Data recovery method and apparatus|
|US6880052||Mar 26, 2002||Apr 12, 2005||Hewlett-Packard Development Company, Lp||Storage area network, data replication and storage controller, and method for replicating data using virtualized volumes|
|US6901481||Feb 22, 2001||May 31, 2005||Stratus Technologies Bermuda Ltd.||Method and apparatus for storing transactional information in persistent memory|
|US6928513||Mar 26, 2002||Aug 9, 2005||Hewlett-Packard Development Company, L.P.||System and method for managing data logging memory in a storage area network|
|US6934826||Mar 26, 2002||Aug 23, 2005||Hewlett-Packard Development Company, L.P.||System and method for dynamically allocating memory and managing memory allocated to logging in a storage area network|
|US6947981||Mar 26, 2002||Sep 20, 2005||Hewlett-Packard Development Company, L.P.||Flexible data replication mechanism|
|US6993523||Dec 5, 2000||Jan 31, 2006||Silicon Graphics, Inc.||System and method for maintaining and recovering data consistency in a data base page|
|US6996672||Mar 26, 2002||Feb 7, 2006||Hewlett-Packard Development, L.P.||System and method for active-active data replication|
|US7007042||Mar 28, 2002||Feb 28, 2006||Hewlett-Packard Development Company, L.P.||System and method for automatic site failover in a storage area network|
|US7032131||Mar 26, 2002||Apr 18, 2006||Hewlett-Packard Development Company, L.P.||System and method for ensuring merge completion in a storage area network|
|US7043504||Apr 10, 2000||May 9, 2006||International Business Machines Corporation||System and method for parallel primary and secondary backup reading in recovery of multiple shared database data sets|
|US7055063 *||Nov 8, 2001||May 30, 2006||International Business Machines Corporation||Method and system for advanced restart of application servers processing time-critical requests|
|US7076481||Mar 24, 2003||Jul 11, 2006||Bmc Software, Inc.||Method and apparatus for logically reconstructing incomplete records|
|US7103586||Mar 29, 2002||Sep 5, 2006||Gravic, Inc.||Collision avoidance in database replication systems|
|US7127478 *||Mar 31, 1999||Oct 24, 2006||Siemens Aktiengesellschaft||Data base for persistent data|
|US7149769||Mar 26, 2002||Dec 12, 2006||Hewlett-Packard Development Company, L.P.||System and method for multi-destination merge in a storage area network|
|US7177866||Sep 17, 2003||Feb 13, 2007||Gravic, Inc.||Asynchronous coordinated commit replication and dual write with replication transmission and locking of target database on updates only|
|US7181476 *||Apr 30, 2003||Feb 20, 2007||Oracle International Corporation||Flashback database|
|US7287078 *||Oct 31, 2003||Oct 23, 2007||Hewlett-Packard Development Company, L.P.||Restoration of lost peer-to-peer offline transaction records|
|US7310743||Jan 31, 2005||Dec 18, 2007||Emc Corporation||Data recovery method and apparatus|
|US7398422||Aug 31, 2004||Jul 8, 2008||Hitachi, Ltd.||Method and apparatus for data recovery system using storage based journaling|
|US7415470||May 17, 2005||Aug 19, 2008||Oracle International Corporation||Capturing and re-creating the state of a queue when migrating a session|
|US7447710||Aug 9, 2004||Nov 4, 2008||Sybase, Inc.||Database system providing self-tuned parallel database recovery|
|US7493347||Jun 2, 2005||Feb 17, 2009||International Business Machines Corporation||Method for condensing reported checkpoint log data|
|US7496787||Apr 29, 2005||Feb 24, 2009||Stratus Technologies Bermuda Ltd.||Systems and methods for checkpointing|
|US7502824||May 1, 2006||Mar 10, 2009||Oracle International Corporation||Database shutdown with session migration|
|US7542986||Mar 26, 2002||Jun 2, 2009||Hewlett-Packard Development Company, L.P.||System and method for maintaining order for a replicated multi-unit I/O stream|
|US7587400||Apr 1, 2005||Sep 8, 2009||Oracle International Corporation||Suspending a result set and continuing from a suspended result set for transparent session migration|
|US7613710||Nov 3, 2009||Oracle International Corporation||Suspending a result set and continuing from a suspended result set|
|US7631018||Dec 8, 2009||Oracle International Corporation||Flashback database|
|US7640275 *||Jan 2, 2004||Dec 29, 2009||Oracle International Corporation||Method and system for controlling recovery downtime|
|US7644093 *||Jan 5, 2010||Siemens Industry, Inc.||Partially embedded database and an embedded database manager for a control system|
|US7693889 *||Dec 28, 2005||Apr 6, 2010||Emc Corporation||Automated backup and recovery for content repository|
|US7739237||Feb 6, 2007||Jun 15, 2010||Gravic, Inc.||Data input routing after failure|
|US7743333||Jun 22, 2010||Oracle International Corporation||Suspending a result set and continuing from a suspended result set for scrollable cursors|
|US7761741||Jun 20, 2008||Jul 20, 2010||Hitachi, Ltd.||Method and apparatus for data recovery system using storage based journaling|
|US7783848||Aug 24, 2010||Hitachi, Ltd.||Method and apparatus for backup and recovery using storage based journaling|
|US7885938 *||Feb 8, 2011||Symantec Corporation||Techniques for granular recovery of data from local and remote storage|
|US7886298||Feb 8, 2011||Hewlett-Packard Development Company, L.P.||Data transfer protocol for data replication between multiple pairs of storage controllers on a san fabric|
|US7953698||May 31, 2011||Sybase, Inc.||Replication system with methodology for replicating stored procedure calls|
|US7979741||Jul 12, 2011||Hitachi, Ltd.||Method and apparatus for data recovery system using storage based journaling|
|US8060780 *||Dec 16, 2008||Nov 15, 2011||Sap Ag||Transient transaction execution with on-demand persistency|
|US8145686 *||Mar 27, 2012||Microsoft Corporation||Maintenance of link level consistency between database and file system|
|US8204864||Jun 19, 2012||Network Appliance, Inc.||Restartable dump that produces a consistent filesystem on tapes|
|US8234473||Jul 12, 2010||Jul 31, 2012||Hitachi, Ltd.||Method and apparatus for backup and recovery using storage based journaling|
|US8392931||Dec 16, 2010||Mar 5, 2013||Hewlett-Packard Development Company, L.P.||Data transfer protocol for data replication between multiple pairs of storage controllers on a SAN fabric|
|US8402147||Apr 10, 2007||Mar 19, 2013||Apertio Limited||Nomadic subscriber data system|
|US8453013 *||Jul 9, 2008||May 28, 2013||Google Inc||System-hang recovery mechanisms for distributed systems|
|US8782085||Apr 10, 2007||Jul 15, 2014||Apertio Limited||Variant entries in network data repositories|
|US8996572||Apr 20, 2012||Mar 31, 2015||Apertio Limited||Variant entries in network data repositories|
|US9092379||Jul 18, 2012||Jul 28, 2015||Hitachi, Ltd.||Method and apparatus for backup and recovery using storage based journaling|
|US9112873||Apr 10, 2007||Aug 18, 2015||Apertio Limited||Alias hiding in network data repositories|
|US9164856 *||Nov 11, 2013||Oct 20, 2015||International Business Machines Corporation||Persistent messaging mechanism|
|US9176772||Feb 11, 2005||Nov 3, 2015||Oracle International Corporation||Suspending and resuming of sessions|
|US9213609 *||Dec 16, 2003||Dec 15, 2015||Hewlett-Packard Development Company, L.P.||Persistent memory device for backup process checkpoint states|
|US9251002||Dec 16, 2013||Feb 2, 2016||Stratus Technologies Bermuda Ltd.||System and method for writing checkpointing data|
|US9354985 *||Sep 11, 2014||May 31, 2016||Huawei Technologies Co., Ltd.||Data processing device and data processing method|
|US9405680||Dec 10, 2010||Aug 2, 2016||Hewlett Packard Enterprise Development Lp||Communication-link-attached persistent memory system|
|US20020059306 *||Nov 8, 2001||May 16, 2002||International Business Machines Corporation||Method and system for advanced restart of application servers processing time-critical requests|
|US20020124202 *||Mar 5, 2001||Sep 5, 2002||John Doody||Coordinated Recalibration of high bandwidth memories in a multiprocessor computer|
|US20020133507 *||Mar 29, 2002||Sep 19, 2002||Iti, Inc.||Collision avoidance in database replication systems|
|US20020144175 *||Mar 28, 2001||Oct 3, 2002||Long Finbarr Denis||Apparatus and methods for fault-tolerant computing using a switching fabric|
|US20030187847 *||Mar 26, 2002||Oct 2, 2003||Clark Lubbers||System and method for ensuring merge completion in a storage area network|
|US20030187861 *||Mar 26, 2002||Oct 2, 2003||Clark Lubbers||System and method for maintaining order for a replicated multi-unit I/O stream|
|US20030187945 *||Mar 26, 2002||Oct 2, 2003||Clark Lubbers||Flexible data replication mechanism|
|US20030187947 *||Mar 26, 2002||Oct 2, 2003||Clark Lubbers||System and method for multi-destination merge in a storage area network|
|US20030188035 *||Mar 26, 2002||Oct 2, 2003||Clark Lubbers||Data transfer protocol for data replication between multiple pairs of storage controllers on a san fabric|
|US20030188114 *||Mar 26, 2002||Oct 2, 2003||Clark Lubbers||Data replication with virtualized volumes|
|US20030188119 *||Mar 26, 2002||Oct 2, 2003||Clark Lubbers||System and method for dynamically managing memory allocated to logging in a storage area network|
|US20030188218 *||Mar 26, 2002||Oct 2, 2003||Clark Lubbers||System and method for active-active data replication|
|US20030188229 *||Mar 26, 2002||Oct 2, 2003||Clark Lubbers||System and method for managing data logging memory in a storage area network|
|US20030188233 *||Mar 28, 2002||Oct 2, 2003||Clark Lubbers||System and method for automatic site failover in a storage area network|
|US20030208464 *||Jan 14, 2003||Nov 6, 2003||International Business Machines Corporation||System and method for optimizing log usage for temporary objects|
|US20030217040 *||Mar 24, 2003||Nov 20, 2003||Bmc Software, Inc.||Method and apparatus for logically reconstructing incomplete records|
|US20040133591 *||Sep 17, 2003||Jul 8, 2004||Iti, Inc.||Asynchronous coordinated commit replication and dual write with replication transmission and locking of target database on updates only|
|US20040148316 *||Jan 2, 2004||Jul 29, 2004||Oracle International Corporation||Method and system for controlling recovery downtime|
|US20040215619 *||Apr 24, 2003||Oct 28, 2004||Bsquare Corporation||System and method for integrating persistent and non-persistent storage in an integrated storage area|
|US20040220961 *||Apr 30, 2003||Nov 4, 2004||Oracle International Corporation||Flashback database|
|US20040254964 *||Jun 12, 2003||Dec 16, 2004||Shoji Kodama||Data replication with rollback|
|US20050028022 *||Aug 31, 2004||Feb 3, 2005||Hitachi, Ltd.||Method and apparatus for data recovery system using storage based journaling|
|US20050096993 *||Oct 31, 2003||May 5, 2005||Salil Pradhan||Restoration of lost peer-to-peer offline transaction records|
|US20050131853 *||Aug 9, 2004||Jun 16, 2005||Sybase, Inc.||Database System Providing Self-Tuned Parallel Database Recovery|
|US20050132250 *||Dec 16, 2003||Jun 16, 2005||Hewlett-Packard Development Company, L.P.||Persistent memory device for backup process checkpoint states|
|US20050216552 *||Mar 24, 2004||Sep 29, 2005||Samuel Fineberg||Communication-link-attached persistent memory system|
|US20050228820 *||Jun 3, 2005||Oct 13, 2005||Michael Soemo||Partially embedded database and an embedded database manager for a control system|
|US20060036616 *||Apr 1, 2005||Feb 16, 2006||Oracle International Corporation||Suspending a result set and continuing from a suspended result set for scrollable cursors|
|US20060036617 *||Apr 1, 2005||Feb 16, 2006||Oracle International Corporation||Suspending a result set and continuing from a suspended result set for transparent session migration|
|US20060059176 *||Apr 1, 2005||Mar 16, 2006||Oracle International Corporation||Suspending a result set and continuing from a suspended result set|
|US20060059228 *||May 17, 2005||Mar 16, 2006||Oracle International Corporation||Capturing and re-creating the state of a queue when migrating a session|
|US20060143528 *||Apr 29, 2005||Jun 29, 2006||Stratus Technologies Bermuda Ltd||Systems and methods for checkpointing|
|US20060159379 *||Dec 22, 2005||Jul 20, 2006||Ab Skf||Bearing arrangement for a medical device|
|US20060184535 *||Feb 11, 2005||Aug 17, 2006||Oracle International Corporation||Suspension and resuming of sessions|
|US20060200454 *||May 1, 2006||Sep 7, 2006||Sanjay Kaluskar||Database shutdown with session migration|
|US20060253502 *||May 6, 2005||Nov 9, 2006||Microsoft Corporation||Maintenance of link level consistency between database and file system|
|US20060277162 *||Jun 2, 2005||Dec 7, 2006||Smith Alan R||Apparatus, system, and method for condensing reported checkpoint log data|
|US20070028144 *||Jul 29, 2005||Feb 1, 2007||Stratus Technologies Bermuda Ltd.||Systems and methods for checkpointing|
|US20070038891 *||Aug 12, 2005||Feb 15, 2007||Stratus Technologies Bermuda Ltd.||Hardware checkpointing system|
|US20070220221 *||May 22, 2007||Sep 20, 2007||Hitachi, Ltd.||Method and apparatus for backup and recovery using storage based journaling|
|US20070226277 *||Feb 6, 2007||Sep 27, 2007||Gravic, Inc.||Data input routing after failure|
|US20070255749 *||Apr 27, 2006||Nov 1, 2007||Vrba Joseph A||Selective extractions of data from large transient data sets|
|US20080034014 *||Aug 3, 2006||Feb 7, 2008||Sybase, Inc.||Replication System with Methodology for Replicating Stored Procedure Calls|
|US20080253403 *||Apr 10, 2007||Oct 16, 2008||Apertio Limited||Nomadic subscriber data system|
|US20080256020 *||Apr 10, 2007||Oct 16, 2008||Apertio Limited||Variant entries in network data repositories|
|US20080256083 *||Apr 10, 2007||Oct 16, 2008||Apertio Limited||Alias hiding in network data repositories|
|US20080256142 *||Apr 10, 2007||Oct 16, 2008||Apertio Limited||Journaling in network data architectures|
|US20090019308 *||Jun 20, 2008||Jan 15, 2009||Hitachi, Ltd.||Method and Apparatus for Data Recovery System Using Storage Based Journaling|
|US20100153778 *||Dec 16, 2008||Jun 17, 2010||Franz Weber||Transient Transaction Execution With On-Demand Persistency|
|US20100251020 *||Sep 30, 2010||Hitachi, Ltd.||Method and apparatus for data recovery using storage based journaling|
|US20100274985 *||Jul 12, 2010||Oct 28, 2010||Hitachi, Ltd.||Method and apparatus for backup and recovery using storage based journaling|
|US20110082992 *||Dec 10, 2010||Apr 7, 2011||Hewlett-Packard Development Company, L.P.||Communication-link-attached persistent memory system|
|US20110087787 *||Apr 14, 2011||Clark Lubbers||Data transfer protocol for data replication between multiple pairs of storage controllers on a san fabric|
|US20150135002 *||Nov 11, 2013||May 14, 2015||International Business Machines Corporation||Persistent messaging mechanism|
|US20150154087 *||Sep 11, 2014||Jun 4, 2015||Huawei Technologies Co., Ltd.||Data processing device and data processing method|
|US20150234615 *||May 5, 2015||Aug 20, 2015||Huawei Technologies Co., Ltd.||Data processing device and data processing method|
|US20150347049 *||Jun 22, 2015||Dec 3, 2015||Emc Corporation||Snapshotting of a performance storage system in a system for performance improvement of a capacity optimized storage system|
|CN100390747C||Feb 20, 2006||May 28, 2008||南京联创科技股份有限公司||Soft co-standby method based on memory database in whole-process real time system|
|WO2004097552A2 *||Apr 8, 2004||Nov 11, 2004||Bsquare Corporation||System and method for integrating persistent and non-persistent storage in an integrated storage area|
|WO2004097552A3 *||Apr 8, 2004||May 18, 2006||Bsquare Corp||System and method for integrating persistent and non-persistent storage in an integrated storage area|
|U.S. Classification||1/1, 714/E11.122, 714/E11.121, 707/E17.007, 707/999.202|
|International Classification||G06F11/14, G06F17/30|
|Cooperative Classification||Y10S707/99953, G06F11/1471|
|European Classification||G06F11/14A12, G06F17/30C|
|Apr 10, 1996||AS||Assignment|
Owner name: TELEFONAKTIEBOLAGET LM ERICSSON, SWEDEN
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:NILSSON, SVEN I.;TODOROVIC, ZORAN;REEL/FRAME:007973/0996;SIGNING DATES FROM 19960329 TO 19960409
|Aug 23, 2001||FPAY||Fee payment|
Year of fee payment: 4
|Aug 24, 2005||FPAY||Fee payment|
Year of fee payment: 8
|Aug 24, 2009||FPAY||Fee payment|
Year of fee payment: 12